@growthub/cli 0.14.4 → 0.14.6

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/templates/prompt-templates/video-generation.md

@@ -1,11 +0,0 @@
-# Video Generation Prompt Template
-
-## Objective
-
-## Prompt
-
-`[subject], [starting state], [motion verb], [camera move], [environment], [lighting], [tempo], [ending beat]`
-
-## Exclusions
-
-## Recommended Controls

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/templates/review-qa-checklist.md

@@ -1,26 +0,0 @@
-# Review QA Checklist Template
-
-## Creative Review
-
-- [ ] Visual direction matches brief
-- [ ] Prompt intent is preserved across variants
-- [ ] No off-brand styling
-
-## Technical Review
-
-- [ ] Correct studio and model used
-- [ ] Aspect ratio / duration / resolution match plan
-- [ ] Inputs and references were attached correctly
-
-## Studio-Specific Review
-
-- [ ] Image: composition and subject integrity hold
-- [ ] Video: motion and continuity hold
-- [ ] Lip Sync: mouth sync and facial stability hold
-- [ ] Cinema: lensing and camera grammar hold
-
-## Sign-off
-
-- Reviewer:
-- Date:
-- Status:

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/templates/self-eval.md

@@ -1,67 +0,0 @@
-# Self-evaluation template — v1 primitive (Growthub, capability-agnostic)
-
-A skill's `selfEval` contract declares **what good looks like** (`criteria[]`) and **the retry ceiling** (`maxRetries`). The agent drives the loop; this file is the pattern the agent follows at every unit of work.
-
-The contract is defined in `@growthub/api-contract/skills::SkillSelfEval` and is **capability-agnostic** — the same primitive applies to code edits, copy drafts, CRM rules, API payloads, asset renders, audit passes, and every other domain. The "unit of work" is the smallest reversible change the skill operates on; concrete boundaries are defined in the kit's own `skills.md` operator runbook, not here.
-
-## Loop
-
-```
-for attempt in 1..maxRetries:          # default maxRetries = 3
-  generate       — produce a proposal (whatever the skill produces)
-  apply          — materialise the proposal against the fork
-  self-evaluate  — for each criterion in skill.selfEval.criteria:
-                     check → pass / fail / needs-confirmation
-  record         — append to project.md (human) + trace.jsonl (machine)
-
-  if all criteria pass: break
-  if attempt == maxRetries: park with needs_confirmation note. Stop.
-```
-
-## What each attempt writes
-
-### Human row — append to `.growthub-fork/project.md` "Session log"
-
-```md
-### 2026-04-23 18:40 UTC · <skill-slug>
-- **Plan.** What you intend to do this unit of work.
-- **Changes.** Material change applied (must correspond to a trace event).
-- **Self-eval.** attempt 2/3 — criterion "<one of skill.selfEval.criteria>" ✗ (<note>). Retrying.
-- **Outcome.** retry-pending / pass / parked.
-- **Next.** What the next attempt (or next session) should pick up.
-```
-
-### Machine row — append to `.growthub-fork/trace.jsonl` (via `appendKitForkTraceEvent`)
-
-```json
-{
-  "ts": "2026-04-23T18:40:00.000Z",
-  "forkId": "<fork-id>",
-  "kitId": "<kit-id>",
-  "type": "self_eval_recorded",
-  "summary": "attempt 2/3 failed on <criterion>",
-  "detail": {
-    "skill": "<skill-slug>",
-    "attempt": 2,
-    "maxRetries": 3,
-    "criterion": "<one of skill.selfEval.criteria>",
-    "outcome": "fail",
-    "notes": "<short explanation>"
-  }
-}
-```
-
-## Hard rules
-
-1. **Never exceed `maxRetries`.** At the ceiling, park with a `needs_confirmation` note and stop. Do not loop forever.
-2. **Record every attempt.** Even the first pass. Continuity depends on a complete journal.
-3. **Keep `project.md` and `trace.jsonl` in sync.** One without the other is a protocol violation.
-4. **Preserve prior entries.** Append-only. Never rewrite history.
-
-## Bound the blast radius
-
-Evaluate at the smallest unit of work the skill operates on, not only at the end of a multi-step run. The retry ceiling applies to that unit — so a bad attempt does not sink the whole job.
-
-What counts as a "unit of work" is **kit-specific** and lives in that kit's `skills.md` operator runbook — not in this agnostic template. A kit picks its own smallest reversible boundary and documents it in `skills.md`; this template does not prescribe one.
-
-The agnostic SDK (`@growthub/api-contract/skills::SkillSelfEval`) does not encode any domain-specific boundaries — it only carries `criteria[]`, `maxRetries`, and `traceTo`. See `SKILL.md::selfEval.criteria` in this kit for the concrete checks, and `skills.md` for the kit-specific unit of work.

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/templates/shot-plan.md

@@ -1,11 +0,0 @@
-# Shot Plan Template
-
-| Shot ID | Beat | Studio | Model | Assets | Camera Intent | Prompt Key | Ratio / Duration | Passes | Review Note |
-|---|---|---|---|---|---|---|---|---|---|
-| S01 | | | | | | | | | |
-
-## Arc Notes
-
-## Asset Dependencies
-
-## Review Gates

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/templates/studio-selection-brief.md

@@ -1,18 +0,0 @@
-# Studio Selection Brief Template
-
-## Requested Outcome
-
-## Recommended Studio
-
-| Option | Fit | Notes |
-|---|---|---|
-| Image | | |
-| Video | | |
-| Lip Sync | | |
-| Cinema | | |
-
-## Why the Winning Studio Fits
-
-## Required Inputs
-
-## Fallback Path

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/templates/visual-campaign-brief.md

@@ -1,24 +0,0 @@
-# Visual Campaign Brief Template
-
-## Project Overview
-
-| Field | Value |
-|---|---|
-| Client | |
-| Project | |
-| Objective | |
-| Primary platform | |
-| Runtime mode | |
-| Due date | |
-
-## Audience and Offer
-
-## Brand and Visual Rules
-
-## Existing Assets
-
-## Studio Hypothesis
-
-## Success Criteria
-
-## Output Package

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/validation-checklist.md

@@ -1,49 +0,0 @@
-# Validation Checklist — Open Higgsfield Studio v1
-
-Use this checklist before considering a package complete.
-
----
-
-## PRE-SESSION CHECKLIST (operator runs before starting the agent)
-
-- [ ] Growthub local Working Directory is pointed at this folder (or Claude Code as alternative)
-- [ ] `.env` exists (copy from `.env.example` if not)
-- [ ] `MUAPI_API_KEY` is set in `.env` (not placeholder value)
-- [ ] `node setup/verify-env.mjs` exits 0 (key is valid and API is reachable)
-- [ ] `bash setup/check-deps.sh` passes (local-fork mode only)
-- [ ] Local fork is running at `http://localhost:3001` OR execution mode is `browser-hosted` or `desktop-app`
-- [ ] `output/` directory exists (auto-created by agent on first write)
-- [ ] Active brand kit exists: `brands/<client-slug>/brand-kit.md` OR using `brands/growthub/brand-kit.md`
-
----
-
-## KIT VALIDATION
-
-- [ ] `kit.json` exists and every `frozenAssetPaths` entry exists
-- [ ] bundle manifest matches kit id and worker id
-- [ ] entrypoint path exists
-- [ ] example brand is public-safe
-- [ ] docs, templates, and examples are all self-contained in this folder
-
-## METHODOLOGY VALIDATION
-
-- [ ] `skills.md` defines strict workflow order
-- [ ] entrypoint points back to `skills.md`
-- [ ] runtime assumptions, provider adapter notes, and output standards exist
-- [ ] visual-production-specific templates exist for all four studios
-- [ ] frame-analysis primitive is documented
-
-## OPERATIONAL VALIDATION
-
-- [ ] local fork inspection instructions exist
-- [ ] browser-hosted flow is documented
-- [ ] desktop-app flow is documented
-- [ ] Muapi auth, submit, poll, and result assumptions are documented
-- [ ] fallback provider extension path is documented
-
-## OUTPUT VALIDATION
-
-- [ ] required output categories all have templates
-- [ ] example files demonstrate the expected contract
-- [ ] naming convention is explicit
-- [ ] review and QA gates are explicit

package/assets/worker-kits/growthub-open-higgsfield-studio-v1/workers/open-higgsfield-studio-operator/CLAUDE.md

@@ -1,286 +0,0 @@
-# Open Higgsfield Studio Operator — Agent Operating Instructions
-
-**Kit:** `growthub-open-higgsfield-studio-v1`  
-**Worker ID:** `open-higgsfield-studio-operator`  
-**Version:** `1.0.0`
-
----
-
-## YOUR ROLE
-
-You are the Growthub Open Higgsfield Studio Operator. You turn campaign goals, brand inputs, and asset constraints into execution-ready visual generation artifacts for a forked Open Higgsfield AI environment.
-
-**You produce:**
-- Visual campaign briefs
-- Studio selection briefs
-- Model selection recommendations
-- Shot plans
-- Prompt matrices
-- Generation batch plans
-- Asset reuse and tracking plans
-- Review notes
-- Platform-ready execution handoff docs
-
-**You do NOT produce:**
-- Vague ideation with no execution path
-- Prompts before confirming studio mode and model path
-- Provider credentials or raw secrets
-- Speculation about repo behavior without checking the local fork
-- One-off CLI automation unless the active environment already requires it
-
-**Your source of truth for methodology is `skills.md`. Read it before beginning any task.**
-
----
-
-## MASTER SKILL DOC
-
-Always read `skills.md` at the start of every session. It defines:
-- Workflow order
-- Required source files in the local fork
-- Studio selection logic
-- Model selection rules
-- Prompt planning rules
-- Frame-analysis primitives
-- Output artifact order
-- QA and handoff standards
-
-If `skills.md` cannot be read, stop and report the error.
-
----
-
-## WORKFLOW — 10 STEPS, STRICT ORDER, NO SKIPPING
-
-### STEP 0 — Environment gate (run before everything else)
-
-Before loading any methodology or brand context, verify the environment is ready.
-
-**Check 1 — `.env` file exists:**
-
-If `.env` is missing, stop and tell the user:
-
-> `.env` not found. Run: `cp .env.example .env` then add your `MUAPI_API_KEY`. Get a key at https://muapi.ai/dashboard
-
-**Check 2 — `MUAPI_API_KEY` is set:**
-
-Read `.env` and confirm `MUAPI_API_KEY` is present and is not the placeholder value `your_muapi_key_here`.
-
-If it is missing or is the placeholder, stop and tell the user:
-
-> `MUAPI_API_KEY` is not set. Open `.env` and paste your key from https://muapi.ai/dashboard
-
-**Check 3 — Key verification (recommended):**
-
-Tell the user they can verify the key before proceeding:
-
-```bash
-node setup/verify-env.mjs
-```
-
-**Check 4 — Local fork (local-fork mode only):**
-
-If the session will use `local-fork` execution mode, check whether the fork is accessible at `http://localhost:3001`. If not reachable, tell the user to run:
-
-```bash
-bash setup/clone-fork.sh
-```
-
-Or switch execution mode to `browser-hosted`.
-
-Do not proceed to Step 1 until the env gate passes.
-
----
-
-### STEP 1 — Read methodology + load brand context
-
-Read:
-
-```text
-skills.md
-brands/growthub/brand-kit.md
-```
-
-If a client brand kit exists, load that instead. If not, start from `brands/_template/brand-kit.md`.
-
----
-
-### STEP 2 — Read runtime and fork docs
-
-Read:
-
-```text
-runtime-assumptions.md
-docs/open-higgsfield-fork-integration.md
-docs/provider-adapter-layer.md
-output-standards.md
-validation-checklist.md
-```
-
-These files define the environment boundary. Do not improvise around them.
-
----
-
-### STEP 3 — Inspect the local Open Higgsfield fork before planning
-
-Before writing prompts or shot plans, inspect the actual working substrate if a fork is available.
-
-Priority source-of-truth files:
-
-```text
-README.md
-package.json
-app/studio/page.js
-components/StandaloneShell.js
-components/ApiKeyModal.js
-packages/studio/src/index.js
-packages/studio/src/models.js
-packages/studio/src/muapi.js
-packages/studio/src/components/ImageStudio.jsx
-packages/studio/src/components/VideoStudio.jsx
-packages/studio/src/components/LipSyncStudio.jsx
-packages/studio/src/components/CinemaStudio.jsx
-electron/
-```
-
-If the user is not pointing at a fork checkout, proceed using the assumptions frozen in this kit and explicitly mark the plan as `repo-unverified`.
-
----
-
-### STEP 4 — Ask the 3-question gate
-
-Ask exactly 3 clarification questions before drafting. Use the highest-risk unknowns:
-
-1. What is the primary output objective: image set, motion clip, lip sync deliverable, or cinematic scene package?
-2. What source assets already exist: brand kit, reference frames, source video, portrait, audio, product stills, or a local mp4 for frame analysis?
-3. What is the delivery constraint: browser workflow, desktop app workflow, or local fork execution with provider adapter access?
-
-Do not generate prompts until these are answered or clearly inferable.
-
----
-
-### STEP 5 — Select the studio mode
-
-Map the job to one primary studio:
-- `image`
-- `video`
-- `lip-sync`
-- `cinema`
-
-Use `templates/studio-selection-brief.md` to document:
-- requested outcome
-- reason for studio choice
-- input asset requirements
-- output constraints
-- fallback studio if the first model path fails
-
----
-
-### STEP 6 — Choose model class and constraints
-
-Read the local model definitions if available, especially `packages/studio/src/models.js`.
-
-Produce a model recommendation that states:
-- studio mode
-- model class
-- endpoint assumption
-- prompt requirement
-- resolution / duration / aspect ratio constraints
-- reference image count assumptions
-- fallback model
-
-Use `templates/model-selection-recommendation.md`.
-
----
-
-### STEP 7 — Build the visual execution artifacts
-
-Write in this order:
-1. Visual campaign brief
-2. Shot plan
-3. Prompt matrix
-4. Generation batch plan
-5. Asset tracking sheet
-6. Review QA checklist
-7. Platform-ready execution handoff
-
-If input footage or a muse video exists, run the frame-analysis workflow from `templates/frame-analysis.md` before the shot plan.
-
----
-
-### STEP 8 — Match the execution mode
-
-Pick one execution path:
-- `browser-hosted` — prompt and asset handoff for the hosted browser app
-- `desktop-app` — prompt and asset handoff for Electron app usage
-- `local-fork` — handoff keyed to inspected source files and provider adapter assumptions
-
-Do not claim the environment can do something the inspected fork does not support.
-
----
-
-### STEP 9 — Log the deliverable
-
-Outputs must be saved under:
-
-```text
-output/<client-slug>/<project-slug>/
-```
-
-Append a deliverable line in the active brand kit:
-
-```text
-- YYYY-MM-DD | Open Higgsfield Visual Package v<N> — <Project Name> | output/<client-slug>/<project-slug>/
-```
-
----
-
-## CRITICAL RULES
-
-| Rule | Meaning |
-|---|---|
-| Env gate must pass first | No `.env` = no session. Key must be set before Step 1. |
-| Read `skills.md` first | No memory-only operation |
-| Inspect the fork before planning | `models.js` and studio components outrank assumptions |
-| Pick one primary studio | No mixed-mode output without explicit transition notes |
-| Model choice must name a fallback | Always provide a second viable path |
-| Muapi is the reference adapter | Do not hardcode it as the only future option |
-| `x-api-key` auth is required | Never invent a different auth contract |
-| Submit -> poll -> result matters | Plans must reflect async generation |
-| Reuse history matters | Check reusable assets before requesting re-upload |
-| Browser and desktop are distinct modes | Call out UI-specific risks |
-| Outputs must be operational | Every file should help an operator execute immediately |
-
----
-
-## REQUIRED OUTPUT ORDER
-
-1. `VisualCampaignBrief`
-2. `StudioSelectionBrief`
-3. `ModelSelectionRecommendation`
-4. `ShotPlan`
-5. `PromptMatrix`
-6. `GenerationBatchPlan`
-7. `AssetTracking`
-8. `ReviewQAChecklist`
-9. `PlatformReadyExecutionHandoff`
-
----
-
-## Governed-workspace primitives (v1.2)
-
-This workspace carries the six architectural primitives every Growthub fork inherits. The contract is capability-agnostic (`@growthub/api-contract/skills::SkillManifest`); kit-specific specialisation lives in `skills.md` above.
-
-1. **`SKILL.md`** at the kit root — the discovery entry / routing menu. Read before `skills.md`.
-2. **Repo-root `AGENTS.md` pointer** — Cursor / Claude / Codex all read the same contract.
-3. **`.growthub-fork/project.md`** — session memory, seeded at init/import from `templates/project.md`. Append a dated entry after every material change.
-4. **Self-evaluation (`selfEval.criteria` + `maxRetries`)** — generate → apply → evaluate → record; retry up to 3; every attempt writes to both `project.md` (human) and `trace.jsonl` (machine). Use `recordSelfEval` (`cli/src/skills/self-eval.ts`); never bypass the fork-trace primitive.
-5. **Nested `skills/<slug>/SKILL.md`** — sub-skill lanes for parallel sub-agents on heavy or narrow work.
-6. **`helpers/<verb>.{sh,mjs,py}`** — safe shell tool layer; promote any inline shell that gets used twice.
-
-Command surface from inside this fork:
-
-- `growthub skills list` — enumerate this fork’s SKILL.md tree
-- `growthub skills validate` — strict shape check
-- `growthub skills session show` — print the current `.growthub-fork/project.md`
-- `growthub skills session init --kit <kit-id>` — (re-)seed session memory
-
-Full user-facing narrative: `cli/assets/worker-kits/growthub-custom-workspace-starter-v1/docs/governed-workspace-primitives.md` (also shipped into any workspace forked from the starter kit).

package/assets/worker-kits/growthub-open-montage-studio-v1/.env.example

@@ -1,42 +0,0 @@
-# OpenMontage Studio — Environment Variables
-# Copy this file to .env and fill in the keys you have.
-# Every key is optional — more keys = more tools available.
-
-# ─── Image + Video Gateway (recommended first key) ───
-FAL_KEY=your_fal_key_here
-
-# ─── Free Stock Media (free developer keys) ───
-PEXELS_API_KEY=your_pexels_key_here
-PIXABAY_API_KEY=your_pixabay_key_here
-UNSPLASH_ACCESS_KEY=your_unsplash_key_here
-
-# ─── Voice & Images ───
-OPENAI_API_KEY=your_openai_key_here
-GOOGLE_API_KEY=your_google_key_here
-ELEVENLABS_API_KEY=your_elevenlabs_key_here
-
-# ─── Music ───
-SUNO_API_KEY=your_suno_key_here
-
-# ─── Video Providers ───
-RUNWAY_API_KEY=your_runway_key_here
-HEYGEN_API_KEY=your_heygen_key_here
-HIGGSFIELD_API_KEY=your_higgsfield_key_here
-HIGGSFIELD_API_SECRET=your_higgsfield_secret_here
-XAI_API_KEY=your_xai_key_here
-
-# ─── Local GPU Video Generation (optional) ───
-# VIDEO_GEN_LOCAL_ENABLED=true
-# VIDEO_GEN_LOCAL_MODEL=wan2.1-1.3b
-
-# ─── OpenMontage Fork Path ───
-# Absolute path to your local OpenMontage clone.
-# Canonical env var: OPEN_MONTAGE_HOME (uniform pattern across all worker kits)
-# Legacy alias: OPENMONTAGE_PATH (still accepted by the CLI)
-# Default when unset: $HOME/OpenMontage
-# OPEN_MONTAGE_HOME=$HOME/OpenMontage
-
-# ─── GrowthHub CMS Node Bridge ───
-# When using CMS nodes for video/image generation, the bridge
-# reads outputs from the hosted execution client. No extra keys
-# needed beyond your GrowthHub session (growthub auth:login).

package/assets/worker-kits/growthub-open-montage-studio-v1/QUICKSTART.md

@@ -1,114 +0,0 @@
-# Quick Start — Open Montage Studio Worker Kit
-
-This folder is your self-contained agent working directory for OpenMontage video production with GrowthHub CMS node integration.
-
----
-
-## Step 1 — Point your Working Directory at this folder
-
-**Growthub local (primary):** Open Growthub local and set the **Working Directory** to this folder in your project settings. Every run you start from this working directory will load the operator agent automatically.
-
-**Claude Code (alternative):** Open Claude Code settings and set the **Working Directory** to this folder. The operator agent loads from `workers/open-montage-studio-operator/CLAUDE.md` on session start.
-
----
-
-## Step 2 — Choose your execution mode
-
-| Mode | What You Need | Best For |
-|---|---|---|
-| `local-fork` | OpenMontage clone + Python + FFmpeg + Node.js | Full production control, all 52 tools |
-| `agent-only` | GrowthHub session or zero-key mode | Quick projects, CMS node generation |
-| `hybrid` | GrowthHub session + local FFmpeg | CMS generation + local post-production |
-
----
-
-## Step 3 — Add your API keys (optional)
-
-```bash
-cp .env.example .env
-```
-
-Open `.env` and add the keys you have. Every key is optional — more keys = more tools.
-
-**Recommended setup order (cheapest first):**
-
-1. **Pexels + Pixabay** — free stock media (free developer keys)
-2. **Google API key** — TTS with 700+ voices (1M chars/month free)
-3. **ElevenLabs** — premium TTS (10K chars/month free)
-4. **fal.ai** — FLUX images + video generation (~$0.03/image)
-
-**Zero keys is fine.** You still get Piper TTS, free archives, and Remotion composition. Or use GrowthHub CMS nodes for generation.
-
----
-
-## Step 4 — Verify your setup (optional but recommended)
-
-```bash
-node setup/verify-env.mjs
-```
-
-Shows which providers are configured and checks for the OpenMontage clone.
-
----
-
-## Step 5 — Set up OpenMontage clone (local-fork mode only)
-
-```bash
-bash setup/check-deps.sh     # Check Python, FFmpeg, Node.js, Git
-bash setup/clone-fork.sh     # Clone OpenMontage and run make setup
-```
-
-This clones OpenMontage to `$HOME/OpenMontage`, installs all dependencies (Python packages, Remotion, Piper TTS), and verifies the tool registry.
-
-Skip this step if you are using agent-only or hybrid mode.
-
----
-
-## Step 6 — Connect GrowthHub (CMS node mode)
-
-If you want to use GrowthHub CMS nodes for video/image generation:
-
-```bash
-growthub auth:login
-```
-
-This authenticates your session so the agent can discover and execute CMS nodes. CMS node costs flow through your GrowthHub billing.
-
-Skip this step if you are using local-fork mode with direct API keys only.
-
----
-
-## Step 7 — Start a new session
-
-**Growthub local:** Start a new run from this working directory. The agent loads automatically.
-
-**Claude Code:** Open a new session with this folder as the Working Directory. The agent reads `skills.md` and the methodology automatically.
-
-In both cases the agent will:
-1. Run the environment gate (check .env, providers, dependencies)
-2. Load the methodology and brand context
-3. Ask you 3 clarifying questions before generating anything
-
----
-
-## What you can ask
-
-Copy any of these into your agent session:
-
-- "Make a 60-second animated explainer about how neural networks learn"
-- "Create a 30-second cinematic product trailer for a smart water bottle"
-- "Make a documentary montage about city life at 4am using real footage only"
-- "Turn this YouTube video into 3 differentiated concepts for my brand"
-- "Create a TikTok-style explainer about CRISPR gene editing"
-
----
-
-## First-run checklist
-
-See `validation-checklist.md` for the full pre-session checklist before your first production run.
-
----
-
-## Creating a brand kit for a new client
-
-See `brands/NEW-CLIENT.md`.